A few of the MAPS lab graduate students will be presenting posters at the 2022 ABCT conference in New York! Details on times you can see them will be posted closer to the date of the conference. For now, check out their abstracts below!
Daniel Brunette: The Efficacy of Four Self-Guided CBT Modules on Depression and Anxiety Symptoms
Even among majority populations in high-income countries, less than half of those experiencing anxiety or depression receive treatment meeting their subjective needs (Mojtabai, 2009). One method of improving access to psychological services is through increasing the availability of low-intensity, brief interventions. To this end, some initiatives, such as Improving Access to Psychological Therapies (IAPT), have assimilated stepped care approaches into their models (Clark, 2011). In stepped care, treatment is provided sequentially with lower intensity interventions, such as computerized CBT and guided self-help, representing the first stage of care with more intensive treatment options at later steps if previous steps don’t result in sufficient improvement. This increases access to services as lower intensity treatments require less therapist time and/or may be performed by interventionists with less intensive training. This is particularly true for self-guided interventions which require minimal to no therapist contact. Additionally, the literature is generally supportive of brief, self-guided interventions. Several meta-analyses have demonstrated that, on average, compared to care-as-usual or waitlist conditions, self-guided interventions significantly reduce symptoms of depression, however, with relatively small effects (Andersson & Cuijpers, 2009; Cuijpers et al., 2011b; Karyotaki et al., 2017). In the current study, we examined the efficacy of four self-guided, CBT skills modules (i.e., cognitive, behavioral, interpersonal, mindfulness) on depression and anxiety symptoms.
Participants (n = 224) were randomized to either a cognitive (n = 61), behavioral (n = 62), interpersonal (n = 47), or mindfulness (n = 54) module. Participation took place over the course of 3 weeks. Each module consisted of three 5-minute educational videos (one per week) on coping skills within the domain of their assigned module. Participants also watched a series of brief review videos to promote comprehension. In addition, participants were expected to practice concepts from the videos by completing 9 module-specific worksheets over the course of the 3 weeks. Assessments of depression (QIDS-SR; Rush et al., 2003) and anxiety (GAD-7; Spitzer et al., 2006) symptoms were completed weekly.
Repeated measures linear regression models were conducted for both depression and anxiety symptoms. In each model, outcome measures were regressed on time (in weeks), while allowing for random intercepts. The results suggested a significant decrease in both depressive, B = -0.75, t(541.57) = -8.94, p < .001, and anxiety, B = -0.63, t(536.19) = -6.25, p < .001, symptoms over time. To determine whether the modules were statistically equivalent in efficacy to one another, two one-sided tests (TOST) of equivalence were performed. Modules tended to be equivalent in reducing symptoms of depression. However, modules did not tend to be equally effective in reducing symptoms of anxiety.
Kassidie Harmon: Differences in BPD features between Black Americans and White Americans
Borderline personality disorder (BPD) is characterized by emotion dysregulation, difficulties in interpersonal relationships, identity disturbance, and self-injurious behaviors (Lieb et al., 2004). Previous research has suggested that racial differences may exist in the presentation of BPD. For example, Haliczer et al. (2020) concluded that race was a significant moderator in relationships between emotion regulation difficulties and BPD features. Similarly, Newhill et al. (2009) found that among a sample of patients with BPD, Black participants experienced greater emotion dysregulation and exhibited fewer self-harming behaviors than White participants.
The literature also suggests that racial stressors can be an important factor in the development of psychopathology symptoms. Experiencing racism is related to depressive and anxious symptoms in Black women (Donovan et al., 2012). Additionally, racial stress and trauma are associated with elevated rates of PTSD within Black Americans (Williams et al., 2018). In the current study, we investigated racial differences in BPD features and the relationship between race related stress and BPD features in Black Americans.
Data were collected from a university sample (N = 159, 77.21% White, 22.79% Black). All participants self-reported demographic characteristics and BPD features (Personality Assessment Inventory-Borderline Scale, Morey, 1991; PAI-BOR). Black participants also completed the Index of Race Related Stress (Utsey, 1996; IRRS), a measure of Black Americans’ subjective ratings of race related stress and experiences of racial discrimination. Compared to White participants (M = 49.60, SD = 12.40), Black participants reported higher PAI-BOR total scale scores (M = 55.47, SD = 12.74; t(156) = 5.44, p = .014). Racial differences emerged within PAI-BOR subscales as well. Black participants reported significantly higher PAI-BOR mood instability (M = 9.28, SD = 3.30) and negative relationship (M = 9.42, SD = 2.61) subscale scores than White participants (M = 8.05, SD = 2.64; t(156) = -2.31, p < .022; M = 8.04, SD = 2.54; t(156) = -2.83, p = .005, respectively). Finally, to determine the relationship between race related stress and BPD features, we conducted a linear regression analysis. Results revealed that for Black participants, race related stress (b = .37) significantly accounted for 13.5% of the variance in PAI-BOR total scale scores (F(1, 34) = 5.29, p = .028).
These results may indicate that Black Americans experience more BPD features compared to White Americans. Alternatively, the PAI-BOR may not measure BPD features equally across race. These analyses also provide support for previous findings that race-related stress may be a contributing factor to psychopathology features. Future research should investigate other potential predictors of psychopathology within Black American samples.
Whitney Whitted: The Effect of Social Media Subtle Communication on Beliefs About Mental Illness Trajectories
Many people with psychiatric symptoms do not seek treatment (Oliver et al., 2018) and when they do, 30-50% dropout prematurely (Roos & Werbart, 2012). One barrier to treatment seeking and uptake is the belief that nothing can be done about depression symptoms. Therefore, those who would potentially benefit may not seek treatment or treatments that rely on patient-initiated behavior (e.g., modifying cognitions and behaviors) may be less effective for those who perceive symptoms as stable and innate. Carol Dweck (1999) coined the terms for fixed and malleable mindsets in the context of studying children’s beliefs regarding intelligence. Specifically, she found that children who viewed intelligence as a malleable feature of one’s personality that can be improved upon with effort performed better academically than those who believed intelligence is fixed and not amenable to change. Since this time, multiple studies have further corroborated and expanded on these findings, providing evidence for malleable mindsets being associated with fewer symptoms of depression and anxiety, lower negative emotionality, and less psychopathology in general (De Castella et al., 2013; Tamir et al., 2007; Shroder et al., 2018).
Given the widespread use of social media to disseminate information about important issues, including psychological health, we sought to understand how the influence of social media communication regarding mental health impacts viewers’ beliefs about their own perceived role in mental illness recovery. Thus, we conducted an experimental study with 321 participants from a large Midwestern university. Participants were randomized into three conditions, fixed, malleable, or control, and viewed a series of fictious tweets. In the fixed condition, participants viewed tweet content presenting mental illness from a fixed mindset perspective (e.g., “I can’t wait for my seasonal depression to be over so that I can get back to my regular depression”). The malleable condition included tweet content presenting mental illness from a malleable mindset perspective (e.g., “user captioned, “I got this,” with a meme that read, “telling those anxious thoughts who’s really in control”). Finally, the control condition did not have any mental health content, but instead was a series of tweets that were generally positive in tone (e.g., “Got my mile time down to 9:20 after it being over 10. I am 10000% not a runner. I cannot breathe but I am so happy!”).
We hypothesized that participants in the fixed condition would be more pessimistic about the possibility of improving mental health trajectories than those in the malleable and control conditions. As hypothesized, one-way ANOVA analyses indicated that participants in the fixed condition believed that depression and anxiety are more chronic (M = 19.79, SD = 4.95) than those in the malleable condition (M = 17.77, SD = 4.90; F(2, 318) = 11.01, p < .001, d = .54). Additionally, participants in the fixed condition believed they had less personal agency over mental illness (M = 10.21, SD = 2.81) than those in both the malleable (M = 8.55, SD = 2.84) and control (M = 9.24, SD = 3.02; F(2, 318) = 9.00, p < .001, d = .59) conditions. These findings suggest that, even in brief exposures, social media content has the ability to impact our beliefs about ourselves and our abilities regarding mental illness, which has implications for both treatment seeking and uptake behaviors.
Dorian Hatch: Investigating the Association between Family Environment Risk Factors and Borderline Personality Disorder
For decades researchers have speculated that certain parental conditions are risk factors for borderline personality disorder (BPD; Chapman, 2019; Crowell et al., 2009). For example, previous research has indicated that parental abuse (Infurna et al., 2016), parental divorce (Plakun, 1991), and parental invalidation (Fruzzetti et al., 2005) are associated with BPD symptoms. Using a large treatment-seeking sample (N = 1145; Female N = 635; Caucasian N = 587; African American N = 370), we used multinomial logistic regression to test whether there was a relationship between BPD diagnosis (versus other personality disorder (OPD) or non-psychiatric control (NPC)) and markers of parental conflict (e.g., parental violence directed at the participant, witnessing intimate partner violence, being raised by someone other than one’s biological parent, and parental separation). We found that reports of witnessing higher levels of intimate partner violence was associated with an increased likelihood of being in the BPD group comparatively to OPD (OR = 0.65, p < .001) or NPC (OR = 0.36, p < .001). Similarly, parental separation was associated with an increased likelihood of being in the BPD group comparatively to the OPD (OR = 0.68, p = .03) or the NPC group (OR = 0.47, p < .001). Interestingly, parental separation moderated the relationship between BPD diagnosis and reports of witnessing intimate partner violence; however, the interaction was associated with a decreased likelihood of being in the BPD group comparatively to the NPC (OR = 1.70, p =.02) and OPD groups (OR = 1.37, p = .03). These findings corroborate earlier research that has posited that certain family environments may be risk factors for BPD. Perhaps our most surprising finding was that parental separation moderated the relationship between BPD diagnosis and violence between parents in the opposite direction of what we expected. In other words, it appears that those with BPD are more likely to report that they had parents who are physically violent towards each other but stay together. These findings also suggest that those who develop BPD report that they observe incongruent behavior from their parents, such as staying in a relationship that perpetuates physical harm. However, these findings should be interpreted cautiously as these data are cross-sectional and retrospective reports, and as such, we are unable to make any causal inferences.
